iPadOS 16
apple.com
apple.com
Just wild to have such an unbelievably powerful piece of hardware that I pretty much only use for youtube and netflix.
You sell a keyboard for it. Let me code on it!
I'd really love to see a native vscode on ipad.
All I need now is a decent terminal for iPad OS AND some kind of USB-C hub to allow charging AND display out the same time.
Do you use Blink by any chance?
I don't know about C#, it's not an ecosystem I swim in at all, but there are multiple python options (including Pyto, the venerable Pythonista and jupyter notebook hosts like Juno & Carnets). There are js environments like Scriptable and Play.js. And it's not Xcode but Swift Playgrounds offers, of course, swift including swift.ui and the ability to redistribute those projects (with some limitations) including submitting directly to the App Store.
I get it; I use an iPad a lot for a great many things. It's not perfect - I have a long list of things I would like to see addressed. Low effort complaints like "let me know when I can code on it" are just that though; low effort. So those who complain "how to code on it?" or about how it exactly match the workflow they use on a completely different platform etc just sound.... whiney.
Low effort noise when some basic infrastructure could enable so many other workflows. I like my iPad, I use it every day, I hope to use more with less friction soon-ish. Yup, same thing every June.
It's the most flexible computer I own. It could be even more so, which I would appreciate, but that depends largely on apple.
For the work I do I can code *locally* but I *don't have to*. Comparing it to a Switch is completely missing what makes them different. Even an attempted reductio ad absurdum should be a little more rigorous than that.
C# is basically "built for IDEs", even VS code can't hold a candle to Rider or Visual Studio - vim and friends are a joke in comparison.
I like iPad form factor and would love the idea of using it as a thin client for development, but unless JetBrains come up with iOS client for remote development (which I doubt) it's a non-starter for me.
Also what's stopping you from charging form USB-C monitor and using it as a hub ?
I remote into a full Windows VM, we use VS 2019.
>>> Also what's stopping you from charging form USB-C monitor and using it as a hub ?
My USB-C monitor can't charge for some reason. It's one of the first 43 inch USB-C supporting monitors.
It wouldn’t be stupid if they would, especially that they are moving towards an architecture of a thin client communicating with the extracted intellij core, even selling remote cloud offerings. Actually, it would likely bring lots of money to them if they come up with a seriously good ipad client.
this isn't about not being able to code on an iPad. that's easy and people having been doing it for about a decade.
this is just about some people still being mad that they don't get root on a machine they buy. I can understand being mad about that, but just say what you mean. don't pretend you can't code on it when people have been doing that basically since day one.
Written by...who knows.
Still not entirely sure what you actually do with it beyond light editing, even with your $5 apps.
- Pythonista is a python IDE for iOS
- iSH Shell is an emulated x86 terminal environment
I still wish they would allow a full dev sandbox on iPad. It can be fully isolated from the root filesystem, I just want to have a terminal, local git repos, and be able to point a text editor at it.
What I don’t understand is why people who have popular apps and decide to stop working on them don’t sell them so someone else can keep them alive. Pythonista is awesome, and lots of people use it. I’m sure someone out there would happily step up to help keep it alive or take it over.
The main difference is in the UI libraries each app provides, which aren't cross-compatible. If you want to run plain text-based Python and work with a REPL, they feel very similar. Pyto ship with Python 3.10, though.
In the past I tried to report bugs to the developer but never got any reply
Honestly the whoLe ‘I wish I could code on my iPad’ thing and complaining kids can’t learn to code on them, has been tired for a very, very long time. Many of these dev environments came out more than 10 years ago now. Being able to develop in swift and publish to the App Store directly for the device for free was announced at WWDC and released last year, but people still keep moaning about it. Pay attention people!
I don't believe there are any well-established general text editors that offers decent git and text editing. Syntax highlighting for a few langs, etc.
And then language runtimes. I do Elixir so I need Erlang plus Elixir. The machine could run those fine but I don't think you would be allowed to ship those via Apple's store.
Plenty of limitations. Kids can probably learn to code on them, depending on what they want and like. The argumenta may be tired but so are the unsolved unsolved problems I feel.
Koder and Textastic are pretty comprehensive coding editors. Working Copy is the premiere git client and works well with both.
I tried running git clone <some reasonably sized project> and ish died.
I am interested in helping out with iSH development, but it isn't really viable at the moment. It's claim to fame is a partial JIT that runs legally despite apple's jit restrictions
This is not a setup for serious development.
Back in real life, however, a Python programming environment limited to "just what it offers in the box" might make some "happy", but it's very easy to argue that it's not serious.
Do you mean the "real life" where UNIX CLI stuff gets cargo culted as the only way a developer works?
there are a few other python environments that support pip install
I agree that the current solutions aren’t great, thus the second part of my comment wishing for a real on-device dev sandbox which would solve many of those issues.
As you say it's entirely an OS limit not a hardware limit.
“Pro” != Developer. There are many other professions
The problem is that if you do all those things on an iPad, you don't own your data. It belongs to whatever walled garden you're working in. Uninstall the app and your data is gone. Stop renting cloud and your data is gone. The company running the app goes bankrupt and your data might be gone.
This is the fundamental difference I see between "mobile" and "desktop" ecosystems. Mobile is designed from the ground up to place you into a much more subservient relationship to vendors. I've described the mobile ecosystem as "feudal computing."
I (honestly) don't understand the appeal of writing code on an iPad. Feels like the Air or Macbook Pro already do that: a keyboard with a screen. I'd feel cramped if I had to develop on something like iPad OS.
I’m going to skip answering and simply say I would like to code on my iPad too
What you call artificially dumbing down others would call optimising for the form factor.
It requires consciously designing everything to work as both a fully fledged computer and content consumption device. That touches everything from product design, internal hardware, thermals, operating systems, frameworks etc.
And what you say about 'consciously designed ...' amounts to nothing meaningful. Have you seen what hardware people have been running Doom on?
So does my watch, my fridge and my car.
iPad's hardware is already a perfectly usable general purpose computer. It is very portable, unlike your fridge, and has powerful processor and high-capacity battery, unlike your watch. The only problem preventing its use as a general purpose computer in a sense of Macbook is a lock by the manufacturer, which prevents loading a proper OS on it, and the absence of hardware specs, nothing more.
And an iPad is nowhere close to the safety-critical nature of the other computers. There is no defence for its locked down nature.
There is almost nothing stopping people from writing up iPad apps for coding things up, except for the fact that there is no market. Apple is no longer standing in the way IMO (stuff like Pythonista exists after all)
Is the context I described responsible for the lack of market? It may be.
I still think that iOS is the best mobile experience out there, but it's
The benefit of the iPad is the touch screen, that's it. The size/weight argument is only valid for those who aren't loading it up with a keyboard cover. I assume everyone talking about coding on it are going to be using the Magic Keyboard.
[1] https://www.theverge.com/2020/4/20/21227741/apple-ipad-pro-m...
https://www.logitech.com/en-us/products/ipad-keyboards/folio...
My iPad has better battery, cellular, etc…
Do I use it to code, no, but that is because I don’t really code anymore. I have used it along with parallels access to remote in to my Air from literally across the world, and run VSCode to troubleshoot some code for a friend. It worked fine. Anymore it’s become my travel machine.
The fact that you can remove the keyboard half is a big benefit to some as well.
It’s also less invasive that way, you can get a laptop stand but it looks ridiculous in coffee shops.
If you are going somewhere and will want to type when you get there, the Magic Keyboard is coming with you. It folds up to act as a case for your iPad during transit... it is a portable device.
iPad OS 16 also adds virtual memory for the first time on an iOS device.
It makes me wonder how much life my 2018 iPad Pro has left. I think it only has 3GB of RAM, which I could see being more important, as Apple adds more multitasking to the OS.
I have also seen that claim, but I have a really hard time believing they didn’t have swap before. Like hell, I’m sure they must have used memory compression as well..
The latter is pretty rare given how much work has been put into notifying apps about memory pressure, making them easily suspend/resume, and not using garbage collection.
Also, sorry to be pedantic but reference counting is a garbage collector algorithm, but I do understand what you mean here (RC needs less memory at the price of not being able to defer work to a concurrent thread)
I would guess that Apple fears that Epic will start using untrusted VMs to distribute Fortnite outside the App Store. Though, honestly, they're not far from being able to do that with a hypothetical WebAssembly/WebGL version already.
I guess I really want some weird hybrid of both OSes.
I have a Mac Mini at home, but lately while doing some prep for interviews (sigh, leetcode), I’ve been going to the library or other places with my iPad and a BT keyboard, and it’s not half-bad to write some code. Unfortunately (or fortunately depending on who you ask) I’ve been mostly writing code on a DigitalOcean instance that I SSH to. That can be trickier if Wi-Fi is shaky where I am since my iPad doesn’t have access to mobile networks and my phone doesn’t do tethering.
Nowadays newer iPads can even hook up to external displays. So form-factor wise, you’re not necessarily missing much, even if you prefer more screen real-estate.
It would definitely be nice to have a native Terminal shell, though.
When I'm home I don't really need the iPad for this (much). There a few use cases where it's clearly better than my MacBook Pro though: the screen is veerry nice outside and I've written more in a hammock than you'd think. Also - airplanes which I am on more often than I'd like to be. an iPad Pro with the magic keyboard is fantastically shallow. It fits on a tray better than anything else.
But you can take it off and write notes with it, pass it around, etc.
If these were your two requirements, you wouldn't need an Air or MBP either.
It's not even about the money, it's the sheer audacity and arrogance that shits me.
Until I can plug a screen and a keyboard to my iPad and write something as sophisticated as Cultured Code's Things on it, there is still ways to go.
The day I can finally code on this thing is the day that I stop carrying a laptop with me.
I gave up on the idea of coding in my iPad when I realized that my iPad Pro, with the keyboard, weighed as much as a MacBook. There was no real benefit after that point, so I stopped trying. :(
It's also only M1-based iPads, though this is a minor distinction.
This is the one I use. It talks about subscriptions, but works for free if you supply your own remote VM/server.
https://apps.apple.com/us/app/servediter-for-code-server/id1...
Just put MacOS on the damn thing already.
Laptops and IDEs are fine. How does anyone possibly want to do anything without multiple screens anyway?
Airplanes are the clearest, but not frequent use case. A 15" MBP is just a bit too big to comfortably use.
Broadly, it'd be nice in more casual settings where I don't want to haul a MBP around.
It's not as good as my laptop and the functionality in every aspect is still very limited. The only unique feature is drawing/notes which might work for artists but paper is good enough for me.
I want to like it but for now it's still a giant iphone with a pen.
It’s not designed for fast (enough) switching or maintaining “hot” data from multiple users simultaneously on the same iPad.
I can’t see that being the reason. The HW difference between the iPad and Mac are minimal.
1: https://mspoweruser.com/nokias-patent-restricts-android-mult...
EDIT: Bottom of article says Apple pays royalties to Nokia to use the patent so maybe they only want to do it for education to keep costs low?
I suspect the author entirely made up the "patent issue" to write this blogspam-like garbage, and that Google's decision to initially restrict the feature to tablets had nothing to do with patents.
This doesn't mean that a 1:1 relationship between a person and their phone / tablet isn't the common configuration - I'm sure it is - but it's fair to say that 'the rest of the world' has had this feature for nearly a decade.
As a sibling comment noted, Apple has developed this feature already in their 'the first hit is free' (aka the Education <sic>) version, so at this point it smells a lot like a plain old vanilla-flavoured marketing ploy.
So if I was Apple, I doubt I’d take the android tablet market seriously. Android mobile phones are viable alternatives but not the tablets.
Samsung's take on Android is, in my experience, dreadful.
I have accounts in 2 countries.
Some apps are only available on one or the other. To install (and some times to upgrade, but not always), I need to switch accounts. When I do that, I lose all my locally downloaded music, podcasts, etc. When I switch back, usually some app doesn’t, like Messages, or even Music.
I usually only notice that says later, by which time I might have missed a lot of things. It’s also a pain trying to sign out and back into one of these apps, nothing is obvious.
My iPad's typically in the living room, if someone want's to quickly look up or show something, this would be ideal.
The regular user would be great for the rest of my family, especially when traveling. We're just not going to get extra iPads for everyone for relatively low time spent on the device.
Unfortunately, in previous cases, these "justification" tweaks have resulted in a strictly worse experience IMO. The transforming cursor on the iPad is distracting and it can be weird how it disappears over icons (resulting in me losing track of it sometimes). The iPad Keyboard is more awkward than a real keyboard, top-heavy, and sold separately of course. I wouldn’t be surprised if the same is true of Stage Manager, although at least it's on the Mac too so hopefully if it turns out to not be that great, it won't be excused as "people not getting it" like all the weirdo undiscoverable gestures on the iPad.
The sad reality though is that it's taken 10 years for the iPad to finally mature into... an arbitrarily different take on OS X? Same chip as the Mac. Works with a keyboard and mouse. Better for drawing but worse for external monitor support. And now basically the same window management we've always had with the Mac. I was always (and remain, theoretically) a big believer in the opportunity the iPad presented, and think it could have gotten to the same place sooner, but bolder, with the actual aim of replacing the Mac (the same way the iPhone replaced the iPod). Where it's at now kind of proves that's always been possible... it just chose to get here meekly instead, not stepping on any toes, always making sure there was an excuse to need to buy both devices.
Universal Control to me is the epitome of that mentality: a supremely technically impressive, but ultimately completely over-engineered and absolutely ridiculous "solution" to a problem that wouldn't exist if the Mac simply had a touch screen. You wouldn't need WiFi and communication between two computers to move your mouse from screen A to screen B if the Mac just had a touch screen. The experience of having a touch Mac that you just... plugged into an external display would be strictly better than the weird world of Universal Control, where you drag an image from Adobe Photoshop for Mac over to Adobe Photoshop for iPad, or sometimes, you drag it from Adobe Photoshop for iPad running on the Mac using Catalyst to Adobe Photoshop for iPad running "natively" on the iPad.
I think Apple truly squandered the opportunity in the last decade to get truly get ahead on "big screen" (aka, non-phone-based) computing. Instead, the field still remains wide open.
> Universal Control to me is the epitome of that mentality: a supremely technically impressive, but ultimately completely over-engineered and absolutely ridiculous "solution" to a problem that wouldn't exist if the Mac simply had a touch screen. You wouldn't need WiFi and communication between two computers to move your mouse from screen A to screen B if the Mac just had a touch screen. The experience of having a touch Mac that you just... plugged into an external display would be strictly better than the weird world of Universal Control, where you drag an image from Adobe Photoshop for Mac over to Adobe Photoshop for iPad, or sometimes, you drag it from Adobe Photoshop for iPad running on the Mac using Catalyst to Adobe Photoshop for iPad running "natively" on the iPad.
There are plenty of use cases for Universal Control that wouldn't be fixed by a touch Mac, though. I use it to control multiple Macs (previously used ShareMouse and before that Synergy, but Universal Control works much more smoothly without a hardwired connection), as well as to use native iPadOS apps from developers that force their Electron apps on macOS with, like Slack and Discord.
Personally speaking I have no desire for touch on traditional MacBooks – I can only see it making sense on a dockable iPad-like Mac.
Yes, the fundamental premise is that the last 10 years would have been used differently to grow the touch space to where it fits in the lineup. Arguably a desktop and laptop are as far apart as a laptop and an iPad. That doesn’t mean they need different OSes. I shouldn’t have to use a completely different OS because I want to use the Apple Pencil. It’s just a peripheral. And yes, peripherals should influence the UI (like how scroll bars appear or disappear depending on whether you have a mouse with a scroll wheel attached). I imagine a “Touch Mac” as a laptop where you could spin the screen around to draw on the go for example. Forget complex touch uses, just being able to fold it back to read like a book or watch a movie is a killer feature. But even the Magic Keyboard for the iPad doesn’t support this transformation, you have to take the iPad out to use it… like an iPad. It’s silly. Similarly, opening up touch to the Mac to allow for things like the Surface Studio, for those who need it, would be great! Instead touch is locked in a tiny screen on Apple’s least-used OS (well, after tvOS I guess).
> Instead touch is locked in a tiny screen on Apple’s least-used OS (well, after tvOS I guess).
If sales figures are any indicator, it's macOS that likely takes the position second-least-used OS. Apple sold 19.1 million iPads in 2021 and only 10 million Macs[0]. Lots of users who'd never consider buying a Mac are in the market for some form of iPad.
[0]: https://appleinsider.com/articles/22/02/23/apples-ipad-domin...
But the last 10 years have been Apple repeatedly disproving this. iPadOS started radically different from macOS... only to have every major inflection point be defined by it moving closer to desktop/macOS. Whether it was giving up on the single-app model to one that recognized that apps needed to run in parallel (first in a "managed window manager" like a tiling window manager you might find in Linux, and later to a more traditional "unmanaged" one like in normal macOS), to shunning away from keyboards and mice entirely, to then only being an accessibility feature, to now being a major add-on that they sell, and even finally adopting key-commands, etc. (I don't know if anyone ever shook an iPad to undo).
I think this position was defensible at the start, but at the very least has had little supporting evidence since. It is indisputable that the evolution of iPadOS has been to move further and further from iOS and closer and closer to macOS. Perhaps there exists some third way (there probably does!), but Apple has proven to be stretched too thin to try anything other than "maybe it should work more like a phone" and "maybe it should work more like a laptop", and the latter has been the consistent, if begrudging, winner.
> The UX requirements just change too much between phone/tablet and laptop/desktop…
I think this has the line in the wrong place. It should be "between phone and tablet/laptop/desktop". The mistake was starting at iOS and having to trudge to macOS over 10 years vs. the other way around (although perhaps it made sense for marketing reasons). I mentioned this before, and Gruber wrote about my thoughts here [1], but the funny thing about the iPhone is that despite being more constrained than the iPad, I feel less frustrated with it. I am impressed with how much I can get done on my phone, I am understanding when it falls short. But the iPad makes it so clear to me all the time that all that's standing in the way of me getting something done is some designer's stubborn reluctance to allow me to do things the "old way". An easy example is writing long form text before there was a good keyboard you could attach. I get why the phone isn't a great place to type, the iPad feels unnecessary for that to always be the case.
> I can kind of see the larger point you're making – it would be nice to be able to directly use an Apple Pencil on a Mac,
I think the best way to think about this is from user workflows first, as I mentioned above it's not even just about the touch aspects. I think the YouTube experience is pretty great on the iPad. Touch here is kind of a bare necessity, I need to pause the video somehow, but it's not a fundamentally "touch experience", it's just a nice form factor for watching videos on the couch. But that experience immediately changes to frustration if I want to suggest the video to someone or have to quickly reply to an email. I'd love to be able to swivel the keyboard out, copy/paste the link, send it in Messages and type something out quickly. Or reply to an email that's longer than a sentence without thinking "ugh, I should really go to my computer for this". This is a point that is bizarrely absent in all these "the devices complement each other" discussions: they work worse separately. I often want to circle something in a picture and draw an arrow, which sucks with a mouse, then type something out, which sucks with a Pencil. The experience is frustrating on both devices, but would be better if I could do both things. And just like a device supporting all sorts of peripherals doesn't mean its a requirement, you could choose to use it as if it didn't have a Pencil or touch as well. There are tons of scenarios that would be way better even in the absolute worst implementation of this: a machine that when in flipped around in "iPad mode" gave you iPadOS, and when the keyboard was out, have it act like macOS (again, I'm not suggesting this is how it should work, but it represents a "floor" of what the experience could be like). Just the weight savings of not needing both devices is great.
> when you try to fudge both models into the same device you get a mess like Windows 8 or GNOME 3/4 which doesn't serve either set of users particularly well.
I've never been impressed with pointing at Linux at Windows as proof that "the other way doesn't work". It is unfortunate that the cost-to-entry on this hardware means we get very little alternatives to compare. And frankly, Windows and Linux don't do that much great in any UI department, that doesn't mean "if Apple can't do it, it must be impossible." If you only had Windows and Linux, you might be convinced that desktop computing was an unsolvable UI problem too...
1. https://daringfireball.net/linked/2020/04/30/tolmasky-ipad-c...
But at the end of the day (in my experience), users fall into one of two buckets- primary use case is productivity (coding, office, photo/movie editing, etc.) or primary use case is consuming content. While there is some overlap between the two users as evidenced by the fact that many people own both and do some of both on each device, it seems inevitable that the devices will continue to converge until a $900 iPad is fully capable of doing 95% of what a $2000 MBP can do and then Apple will have a real problem on its hands as people stop buying the more expensive devices and the cheaper ones last much longer.
We are watching the innovator’s dilemma play out inside a single company. Interestingly one of the tell tale signs a disruption is underway is a hybrid use case where people use a costlier version of a product along side a cheaper version because the cheaper version does certain things better than the costly one.
For professional use, iPad OS will always be a compromise until it's fully integrated with MacOS. And it baffles me that Apple is trying to market iPads to professionals, yet their unwilling to take that step.
One should note that the changes Microsoft initially made to support the Surface (and touchscreens in general) so deeply compromised their desktop OS's design and functionality that many had to be reverted for the next OS release; many other "features" have continued to be among the biggest usability pain points for the ecosystem.
IMO it is incredibly well implemented and thought through. And it works amazingly well and serves a clear purpose for the medium which is iPadOS with all of its design details.
In the user interface/interaction world, I'd go as far as calling it genius.
This video explains in more details the thought process behind the design decisions. https://developer.apple.com/videos/play/wwdc2020/10640/
While we're on this train, the new spotlight button on iOS just seems like worse "Just Type" from webOS.
Its okay to feel like this GUI doesn’t work well for your needs, but that doesn’t mean it was “arbitrary”.
My guess is non-pro laptops will basically be iOS, and then there will be a MacOS Pro, which unlocks some capabilities for developers, but all in a unified code base.
First step was to share processor architecture and work in the ability for iOS apps to run in MacOS. But iOS is very much converging into a replacement for MacOS. And just like iPhone iOS strips out features you unlock with iPadOS, same for the iOS-ification of MacOS.
The whole business case for the iPad falls apart the minute it can install and run arbitrary code. At that point, the iPad shifts from being a large form-factor iPhone to being a multi-modal Macbook. Which is to say, it goes from a device that compliments an existing device to one that replaces and existing, more expensive device.
The reason professionals use Macbooks is because users still need to install and run software that's not sanctioned by Apple. And users are willing to pay a decently large premium to do so.
I'm sure Apple would love to lock down Macbooks the same way they did for iDevices, but if they do that, they would end up killing the Macbook market overnight.
Btw, this doesn’t mean that iOS MacOS will be unusable by devs. I’m sure they have already roadmappped what a pro iOS looks like, for MacBook users. Eg terminal, compiling, code running, etc. But might be more locked down from where you can install .apps from.
Apple straight up does not want people using iPads to do laptop things. They don't want people substituting their $999 Macbook Airs with $329 iPads.
Disagree. Just look at what they are releasing in iPadOS 16. Apps can now float over each other. Team iOS is coming for Team macOS, they are closing in fast. They are developing a UX & app architecture that can easily match consumer expectations around real physical laptops, and there will be further differentiation when it hits Macbooks in a few years time.
With iPadOS 16, for 95% of non-pro users, with a keyboard attachment, there is very very little differentiation between a iPad purchase and a Air purchase. You pretty much own a laptop at that point. Yes, there are obvious performance differences, ports, form factor, battery, etc. But think through the mind of a consumer user, the OS capability differences are razor thin.
If they limited the pro software to the Pro hardware, the iPad vs MacBook prices would be nearly identical.
If Apple allows exceptions to the iOS sandbox for developers, it will very rapidly be abused by users who don’t understand the implications of what they’re doing.
You can see this in the many just-jailbreak-your-iphone-to-get-this-cool-feature videos all over youtube. Almost no-one outside of developers understands what it really means to disable these guardrails and the headache for Apple when it starts going wrong would be enormous.
Grandma: “I let my grandson borrow my phone but now it has this screen that says I need to pay money to decrypt my files how do I fix it?”
Apple store staff: “That appears to be malware, madam”
blank stare
Apple store staff: “I’m afraid there’s nothing we can do”
Grandma: “But the ad said it was secure.”
etc.
The argument that people are too stupid to know how to properly use a computer is just a handy argument their marketing team uses because protecting little old ladies from evil hackers sounds better than the truth.
Also as someone who's grandmother started her own software company and was a programmer since the days of punchcards, I find these "what about grandma" appeals very repetitive and kind of insulting.
"Grandma" doesn't mean "grandma" in these discussions, it's a placeholder for "people not driven to learn the details of all of their tools". Which is most people. Some people do "people" or "animals" or "geology" and not "technology". Your grandma sounds ace, btw.
Your grandmother sounds like a genuinely very impressive person, given that it would have been a much more difficult career to forge as a woman in the timeframe from punchcards forward, even up to the present day.
Respectfully, I don't believe you are insulted. I think you know what was meant by the above, and chose to be insulted.
Your overall argument is also not wrong, but kind of irrelevant. This is HN - everyone here is a tech enthusiast. We are not Apple's target market. Removing the guardrails would cost them revenue and headaches.
Whether or not companies should be made to allow arbitrary software to run on devices is a different question entirely, has no clear and simple answer, and I'm not sure who would have the authority to make that happen.
I really don't think so. The Mac has a different history and use case and is heavily used within Apple, among other important places.
The incentive for Apple to do this would be what: do extract a little more revenue from their smallest-volume OS? As you say, part of its use case is the fact that you can just download and run code on it. That's a small and decreasing percentage of users, but those users are important too.
I'm glad there's no sideloading on my phone and ipad (of course I can always run my own code when I want to). It means my parents have stable platforms to use. And I'm extra glad that restriction doesn't apply to my mac.
I like how the design language of the watch is influencing the ios and macos devices. Unlike the stupid experiments on the phone, there isn't a lot of room on the watch for "flat" interfaces: space is at a premium and contrast is an important factor.
I'm also glad that for almost the first time new features are rolling out to i devices and macos at the same time. Usually it was mobile first, then grudgingly macos.
I think that's only part of the picture. Yes it would boost their services business, but Macbook/iMac users are a small part of the picture these days. It's smarter from a technology centralization standpoint. Why have almost all your products centered around iOS codebase, and then your laptops/iMacs (small % of total devices) on legacy OSX. A better strategy would be, use a single OS codebase, and add layers of functionality the bigger / more pro a device gets. Makes it way easier to advance the software for your entire product line.
Btw, expanding iOS to Macbooks/iMacs does not mean throwing pro / dev users under the bus. They can think of a way to make "iOS-macOS" appealing to devs, by simulating some of the OSX features, like terminal and the ability to write software on-device.
That is the strategy that they have used since the original phone announcement (Jobs touted it on stage). Supposedly there were two competing phone teams, one built on a mac os X base and one on the ipod s/w. I don't know any of the people involved so I can't say how true that is, but it feels a bit weird to me, like the kind of thing that would have been decided pretty early in the process.
In any case there's a large percentage of common code base between the oses on phone, ipad, watch, tv, and mac (and a bunch of embedded devices too, like the usb-c and lightning dongles for HDMI -- yes you can open a console to those devices if you want).
What I think is great is that a bunch of ios frameworks now work/have equivalents on the macOS. I'm actually pretty happy for the same capabilities to be roughly available across platforms: back in the 80s the hardware I used (lispms mostly) was pretty much all a computing "window" into a common underlying sense of computation. I use the apple devices because they are getting pretty close to that: I can just pick up the device that has the most convenient interface for task X, or if that's not possible, still do task X with the device at hand.
Having said that what sucks is when the mac application is just a lazily ported ios app that doesn't fit the mac UX metaphors. But this release looks like a good step forward in that regard.
Maybe, but its better to think holistically.
iPad hardware sales still trail Macs. An argument could be made that iPad hardware sales + iPad-sourced App Store sales does exceed Mac hardware sales + Mac-sourced App Store sales, but it legitimately is not by much. 2022Q2 iPad hardware sales were ~$7B; Mac sales were ~$10B; total Apple Services revenue, including App Store, on iPhone+iPad+Mac, Music, TV+, etc, was ~$19B.
That context emphasizes your statement: Apple would (maybe) love a reality where the Mac were as locked down as iOS, but achieving that reality would torpedo their Mac sales. But, we need to go a step further: it would torpedo their Mac sales as it has their iPad sales.
To me, the most frustrating thing about being a long-time Apple user (and at times, a fan) is watching their spirit of "disrupting themselves" disappear. They systemically can't do it anymore.
A few days ago I was playing around with python on a windows machine; I say "python3 thing.py" in a terminal; it says its not installed; and immediately opens the Windows App Store to install it. I gotta say; that workflow was the most impressed I've been with a computer in four years (and it was immediately paid back by trying to search for "Spotify" in the windows start search and it returning a website as the first result, but I digress). That's self-disruption; the idea that doing the unobvious thing from one revenue-line perspective can lead to % upticks in other revenue lines that net higher total revenue. The Windows Store sucks for buying apps; but doubling it as a dev-focused package manager? And WSL is a cost-center for Microsoft, but jeeze... does a Surface Laptop make sense for me? Maybe.
Apple constantly says "we want it all, its our way or no way" then cracks start appearing in that peerless veneer and out those cracks surge billions in revenue. Other companies (sometimes) say "ok, we're not great at that, how can we work around that weakness to still drive revenue in what we're great at." Microsoft sucks at building a phone; so bring the services to the phone people already have. They suck at building an App Store; but how can we still make that App Store useful, if not profitable? Their consoles have always been in last place; so how can we deliver the main revenue driver, the games, to more people (cloud streaming)? The list goes on.
I'm not asserting that Microsoft is a better company, whatever that means, I'm just asserting that Apple needs to learn something from them, because at this point the two have switched places. Apple today is acting like Microsoft twenty years ago, and vice-versa, and someone at Apple has to be recognizing that iPhone rev growth is slowing, iPad rev is flat, M1 gave the Mac a little boost but it's been slow for 15 years, their most successful new product of the past, like, decade, were headphones. Momentum is great when its with you, but its apt to decay, and once it starts going the other way its a bitch to turn around.
iPads with a keyboard cost more than a Macbook Air
Apple wants you to buy more devices to fill gaps that another one doesn’t support.
I can also make regular calls from my Apple Watch.
I'd have to see it in person, but it looks like "stage manager" occupies a lot of screen real estate for something that is supposed to improve productivity, and it's not clear to me how it plays with iPadOS's existing "split view."
I like my M1 MacBook Pro for programming but I other wise usually use my small or large iPad Pros. I like Linux for dev, using a few application specific VPSs so eventually dropping use of laptops will happen for me. Mosh/tmux/EMacs is usually my best setup.
What I would've liked to see were more splits in the multitasking layouts, so you can show more apps simultaneously, while also maximising the screen used for content.
Not even more developments on the "desk-top class apps", the default apps are still not Mac-parity. On the API side of thing, no API for generic background tasks (like hosting a server for instance).
It's just a niche product and always will be. Making it more like a macbook...just makes it a macbook.
- Lenovo X12 tablet
- Asus ROG flow Z13
- HP Elite X2 (G4 or G8) (what I use)
- Microsoft surface pro (touch support is finally coming for the 7+ and 8)
Will you get good battery life or a fault-free experience? No, it isn't for everyone. Tablet support might break with new releases. But it's been reasonable enough to use as a my only computer for a year after being a died in the wool Mac user, and inspired exploring more ergonomic forms of mobile computing: https://www.reddit.com/r/ErgoMobileComputers/comments/s6k1qr...
Where do I sign up?
I made a collage about my frustrations with iPad a couple years ago ( https://imgur.com/a/CQwApt8 from https://news.ycombinator.com/item?id=22627889 , 2020) and since then:
- One of the accessories I showed in there, gnarbox , a second computer for ipad photo workflows, went out of business and supposedly the app required to make it's main features work is no longer available: https://petapixel.com/2021/12/29/gnarbox-may-be-dead-so-buye...
- Ish still requires a hack to run in the background: https://github.com/ish-app/ish/wiki/Running-in-background (maybe things are changing now with today's announcements?)
- UTM has a number of ios-specific limitations: https://github.com/utmapp/UTM/wiki/Known%20Issues#ios-specif...
- On a couch: https://www.reddit.com/r/ErgoMobileComputers/comments/tcwep0...
- In a car: https://www.reddit.com/r/ErgoMobileComputers/comments/sb4cl5...
- Lying down: https://www.reddit.com/r/ErgoMobileComputers/comments/sh36kr...
- Sitting at a desk: https://www.reddit.com/r/ErgoMobileComputers/comments/s6k1qr...
In the above scenarios I no longer need to use my eyeglasses while doing this as the display is a bit closer to the face of myself as a nearsighted person. And it's enabled a number of reading tasks (books, papers) without carrying a second device.
The limitation is artificial, because the hardware is already capable today. I get why Apple keep doing it this way, but I don't get why people keep defending Apple.
My second monitor has been collecting dust for the past year since switching to a tablet PC, and if really needed consider upcycling a second small tablet to be a second display.
My comment is more about the fact that I can stay in the same device while switching mode, as opposed to switching between the Mac and iPad or whatever. My iPad is mostly unused, I only bought it because I thought sidecar was cool.
I’m also excited about lookup being made better. It seems like an OS feature that could be much better if someone just spent a year trying to come up with new things to look up and adding them. Unit conversion is excellent.
As a result I use all of Notes, Notability and GoodNotes each of which has some capability I use and can't get out of the other two. A terrible situation, though I guess I should be glad there is not a monoculture!
I would love it if the whiteboard fixed these problems, and even better if the fix migrated into Notes.
Both distros build an "immutable system," but in very different ways. Keep in mind that both of these effortlessly deal with rollbacks.
Nix
With NixOS you declare your entire OS in a script, something along the lines of Ansible or Terraform. This can even go as far as configuring your user settings, with dotfiles, gsettings, or various other things (often the modules will expose settings in the native nix language and write out the yaml/toml/json/whatever else that is required). The idiomatic way to do this is to use the built-in NixOS configuration (under /etc/nixos) to set up system-level things: mounts, drivers, users, system-level packages (e.g. greetd+sway), and things that change rarely. You then use a project called home-manager to manage everything inside your user configuration (including applications you use), which itself uses nix. By separating it like this, I can sync my entire experience between my laptop and desktop with Git.
I am currently flighting using a separate "nix flake" for both, which allows you to pin versions of packages (with a lockfile). It also allows you to easily pull in other repositories. It hasn't really taken off yet, and the NUR (Nix User Repository, analogous to the amazing Arch User Repository) is still in infancy. I'd offer up my nix configs as an example, but I am currently in the "make it work" example. I have been yoinking several great ideas from this fantastic nix repo: https://github.com/ymatsiuk/nixos-config
The main challenge with Nix is that it doesn't have an FHS: there is no `/usr`, `/bin`, and what you would typically expect. The advantage here is that conflicting dependency versions are not a problem. The problem is that you need to either build any binaries yourself, or wrap them in an FHS helper.
Nix has a virtualenv system `nix develop` and it's very powerful, especially for teams.
Silverblue
Imagine having a 2-layer container. One contains all the system stuff and is read-only, the other contains your home directory (basically). You can swap out the bottom/system layer at any point to upgrade the system without affecting whatever the user has done. Silverblue "merely" does this efficiently. The problem is, how do you now install packages? Flatpak and Toolbox (or Distrobox if you prefer, which I do).
I have been genuinely amazed at how much stuff now works in Flatpak. You'll get 90% of what you need from there. Distrobox is how you'll approach everything else. It's simply a container with complete access to the host, that also behaves like the host (e.g. the home dir is mounted). It uses containers as a layer, not a sandbox. You do all your development in that and use the container's native package manager to install what you need (apt, dnf, whatever). I actually configured `distrobox enter` as the default shell in my terminal emulator, you should rarely need the host shell.
You can also install packages to the system, but this is strongly discouraged. The only packages I layered in were fish, starship, and a custom kernel. Updates take zero compute (unless you are using a custom kernel, but it's not much), are are applied by restarting. That's also a major disadvantage, if you layer in packages (i.e. by not using Flatpak or Distrobox) you need to reboot in order to use them - it could be argued that a bad practice should involve friction.
Silverblue has done the craziest thing I've ever seen an OS do: I rebased to its sibling distro Kinoite (KDE-based) in almost exactly the time it took to download the system layer, plus the time it took to reboot - so about 2 minutes.
dotfiles are your problem in Silverblue.
I'm currently using Nix, but I am carrying over Distrobox to it and using that instead of `nix develop`.
HTH
As a non-apple user, forgive me for being skeptical but it sounds like (and I'm probably missing something) you're saying one needs an Apple tablet specifically for the grand task of drawing on a touchscreen. Or maybe you're using the word iPad to mean tablet in general and I'm reading too much into the comment?
Aside from an iPod a bunch of years ago, the iPad was my first Apple platform product. Last year I switched from a Google Pixel to an iPhone, so now I have two Apple products. The phone is fine but I was pretty happy on the Pixel too. The iPad though us my favorite computer to use.
My 2020 iPad Pro has an A12Z and 6 gigs of RAM, gating off access to the new stuff completely feels very artificial.
I could do more multitasking on a 486SLC/33 with 4MB of memory.
> My 2020 iPad Pro has an A12Z and 6 gigs of RAM, gating off access to the new stuff completely feels very artificial.
I have that iPad Pro as well, and I'm infuriated. Looking at the Surface line to replace it. Apple's strung me on too long with the iPad Pro.
The ARM and x86 don’t have the same memory barriers so MP is different, but I’m not sure that’s what you’re getting at (it’s an instruction semantics issue non memory controller).
I used to think that Intel had the edge in sustained multi core processing but now I’m not so sure.
The iPad is cool, portable and has a nice screen, but iPadOS still holds it back IMO.
> UTM is supported on iOS 11, 12, and 13 for non-jailbroken devices through sideloading. UTM requires a jailbreak to use on iOS 14.
Remember, even arbitrary code execution is officially not allowed yet (yes there are exceptions)
OTOH, I think Apple's priorities might well give us a calculator app first.
> “There’s some things that we have not done because we would want to do something really distinctly great in that space,” the Apple VP explained. “We want to do it when we can do it really, really well. And we honestly just haven’t gotten around to doing it great.”
Wtf. The UX of a calculator doesn't need to be reinvented. I just need a freaking calculator on my device that I can use. I'd rather have something.
For simple calculations you can use home screen search.
On the other hand, did it really have to take 15 years to get here?
We should really ask WhatsApp at this point :P
Ridiculous. So it's SMS or some other app with iOS users.
Hello Apple? Your market isn't only the USA.
If SMS was to go away, and if RCS wasn't used, then the default will end up being something like Facebook Messenger, which is not desirable since they also own Whatsapp. 1 corporation should not have line-of-sight into the metadata of the majority of the planet's messaging, and possibly also the internal content.
RCS as implemented in Google Messages is not bad. It does include encryption (in Google's implementation) and the other features you'd expect. There are issues, such as AT&T S22 users because AT&T decided to make that device use their own server and not interoperate with the other server. But bumps like that aside, I've tried it out with some people and it works fine and because every phone has a unique phone number, you don't have to establish whether they are a Facebook friend or have the right app first to use it. If it's blue, they can be reached over RCS, and if it's green, then it's off to the SMS races.
for those who are interested
> Workaround: Use a device with an A12 or A13 chip.
Okay.. Another $2000 sure
This is the last piece that I need to make the switch to my iPad Pro only.
It will also be the reason I drop my trusty Pro 11 for a new M2 Pro
Finally! 12 years after the first iPad, Apple released a native Weather app which doesn't point to a website with ads.
Next year they should release a calculator.
Federighi said it would be very easy for Apple to just scale the iPhone Weather app to the iPad, but that's not something the company wants to do, instead looking to create an app that makes use of the larger screen of the iPad in every possible way
[1] https://tech.hindustantimes.com/tech/news/apple-exec-explain...
Either they explicitly didn't want to make one for some reason or they just didn't care. If they actually wanted to make a weather app they could have done it.
There have always been far more important things to work on than an iPad weather app. In this case they happened to get it for free with their Dark Sky acquisition.
For example, the Xsnow accumulation on the top in-app widgets! I wonder - if you shake the iPad, will it clear away?
Why it didn't come with a weather app at release in 2010 I'm not sure, but why it hasn't come in the past few years is almost certainly because they were building it with their new in-house weather API.
> "Weawow is sustained by user donations."